There is a need for the development and evaluation of diagnostic instruments suitable for daily use in primary care offices that can improve recognition rates of psychopathology. The objective of this study is the comparison of the German version of the Primary Care Evaluation of Mental Disorders (PRIME-MD), a short structured diagnostic instrument, with the Composite International Diagnostic Interview (CIDI) and to gather some information on the usefulness of the PRIME-MD. Seven hundred and four patients were assessed three times, once using the physician's clinical judgement, subsequently, administering the PRIME-MD, DSM-IV version and finally, with the CIDI. The CIDI was administered on a different occasion within 1 week after the PRIME-MD evaluation by independent interviewers over the telephone. At the end of the study, the participating physicians answered a few feedback questions on the usefulness of PRIME-MD. Sensitivity (0.73), specificity (0.67), overall accuracy (0.70) were good for any psychiatric disorder. According to the diagnostic categories of mood disorders, anxiety disorders, eating disorders, and alcohol-related disorders, the sensitivity of PRIME-MD was good (0.67-0.80). For somatoform disorders, sensitivity was poor. Specificity, accuracy, and negative predictive values were good-to-excellent for all diagnostic categories. Kappas range from poor (0.12 somatoform disorders) to satisfactory (0.62 alcohol related disorders). Average time for administering PRIME-MD was 11 min. The results demonstrated that the German version of the PRIME-MD may be useful in primary care settings to enhance recognition of mental disorders by primary care physicians, even without being formally trained in diagnosing mental disorders, (C) 2000 Elsevier Science Ltd. All rights reserved.
